Field-Free Magnetization Switching in A1 CoPt Single-Layer Nanostructures for Neuromorphic Computing

Abstract

Current-induced field-free magnetization switching in single-layer films has been widely investigated for the application of spin−orbit torque (SOT) drivers in low-power, high-density, and nano-sized memory. In this paper, we report field-free SOT switching with threefold angle dependence in A1 disordered singlelayer CoPt using the MgO (111) substrate. It is found that the magnetization could not switch in the absence of an external inplane magnetic field when the current was flowing in the direction of the mirror symmetry high-symmetry axis ([11−2]). While along the direction of the low-symmetry axis ([1−10]) which destroyed the mirror symmetry to the greatest extent, field-free switching could be performed, and the switching ratio was up to 30%. This was mainly attributed to the composition gradient-induced in-plane inversion asymmetry and the low symmetry at the interface of Co/Pt broken out-of-plane inversion. Furthermore, CoPt devices with nanoscale thickness exhibited stable multistate magnetic switching behavior without an external magnetic field, and simultaneously nonlinear synaptic characteristics were obtained. A neural network was established, in which the synaptic weights between the hidden layer and the output layer are updated by using the resistance state of SOT synaptic devices, and the network could achieve about 90.54% recognition accuracy when applied to MNIST's handwritten digital dataset. This work confirmed that a field-free SOT switch could be realized in single-layer CoPt, which provided theoretical guidance and data support for spin devices.

Field‐Free Memristive Spin–Orbit Torque Switching in A1 CoPt Single Layer for Image Edge Detection

“…The non-linear synaptic characteristics of CoPt-based SOT devices can be used to simulate neural networks and identify handwritten numbers. [12,33] However, the hardware implementation of neural networks has rarely been reported in the relevant literature.…”

Neuromorphic Computing with Emerging Antiferromagnetic Ordering in Spin–Orbit Torque Devices

“…This can be attributed to massive parallel processing and high energy efficiency . Several reports have suggested that SOT-based devices could be used as artificial synapses for NC. ,,− Yadav et al experimentally constructed LTD/LTP responses for the Pt/Co/SiO 2 SOT system and utilized it as an artificial synapse for NC . Kurenkov et al realized an artificial neuron and synapse in an AFM/FM-based SOT device for NC .…”

“…The current-induced SOT switching loop (Figure d–f) exhibits several intermediate magnetic orientations. This multistate switching, or memristive behavior, is crucial for neuromorphic computing. ,, We investigated these characteristics by a sweeping pulse current under H x = −300 Oe, as detailed in section S7. In addition to the switching induced by current-driven SOT, we observed an analog change in the magnetization of the Co layer when the devices were exposed to a pulsed electric current, as shown in Figure .…”

“…In addition to the switching induced by current-driven SOT, we observed an analog change in the magnetization of the Co layer when the devices were exposed to a pulsed electric current, as shown in Figure . This trait could potentially be used as a synapse ,,− ,,− for information transmission in high-efficiency NC applications. Figure illustrates the neuromorphic characteristics of NiO-based SOT devices under an applied pulse current.…”

“…Hereafter, this scenario is termed ideal . The second scenario was implemented by using an experimentally constructed synapse based on 150 R Hall states (Figure d–f) for synaptic weight , and fitted sigmoidal functions (Figure b) as the hidden layer activation function . The second scenario is experimental and therefore termed exp .…”